Abhishek Vivekanandan

Abhishek Vivekanandan

Research Fellow at FZI Forschungszentrum Informatik

Karlsruhe Institute of Technology

Biography

As a Research Fellow at Forschung Zentrum Informatik and KIT, my current endeavors are centered on addressing the multifaceted challenges associated with the deployment of automated vehicles. My primary objective is to enhance safety measures, a pivotal component in the deployment process and a substantial barrier to attaining verifiable safety standards. With a robust background in the development and execution of deterministic characteristics that align with industry benchmarks, I am committed to ensuring the safe and efficacious deployment of automated vehicles on a large scale.

Interests
  • Machine Learning
  • Highly Automated Driving
  • Perception and Motion Systems
  • Automotive Software Systems
  • Safety for Automated Vehicles
Education
  • PhD in Artificial Intelligence, 2021 - Present

    Karlsruhe Institute of Technology

  • MSc in Computer Science, 2018

    Technische Universität Chemnitz, Germany

  • B.Eng in Electronics and Instrumentation Engineering, 2015

    Anna University, India

Experience

 
 
 
 
 
KIT
PhD Student
KIT
Jan 2021 – Present Karlsruhe,Germany

Development of Plausiblity methodologies for Deep Neural Networks: Towards Functional Conformance

  • Plausiblity Verification for 3D Object Detectors using Energy based Optimization: Framework to reduce False Positives using cross sensor validation.
  • Knowledge Integration Plausible Motion Forecasting: Predicting conformal trajectories with kinematic and environmental priors.
 
 
 
 
 
FZI
Research Scientist
FZI
Jan 2019 – Present Karlsruhe,Germany

Advancing functional safety for deploying Neural Networks in Autonomous Driving.

  • Created a motion prediction model that forecasts trajectories of dynamic actors around the ego vehicle, leveraging fused LIDAR point clouds and HD maps through a novel sensor fusion architecture.
  • Pioneered a multi-modal cutin detector using a CNN and Attention-based model to anticipate lane merges of surrounding vehicles by fusing RADAR and Camera data.
 
 
 
 
 
Volkswagen AG
Master Thesis
Volkswagen AG
Jan 2018 – Oct 2018 Wolfsburg, Germany

Model Uncertainty estimation on Semantic Segmentation Network with a real time deployment on Nvidia Drive PX2 for Autonomous Vehicles.

  • Parallelization of MC Dropouts for real time sampling of Uncertainty.
  • Custom layer optimization of Network model using TensorRT optimizer.
  • Deployment of optimized model into Drive PX2 for real time segmentation.
 
 
 
 
 
Volkswagen AG
Internship
Volkswagen AG
Jun 2017 – Jan 2018 Wolfsburg, Germany

Worked on,

  • Design of Segmentation Network architectures for Deep Neural Network.
  • Development of training Pipeline for DNN using TensorFlow.
  • Uncertainty Estimation of DNN models resulting in a boost in performance of the pixel wise classification accuracy.
  • Benchmarking of IBM cluster GPUs for training performance evaluation.

Recent Publications

Quickly discover relevant content by filtering publications.
(2024). Scene-Specific Trajectory Sets: Maximizing Representation in Motion Forecasting. PreprintArxiv 2024.

PDF Cite DOI

(2023). KI-PMF: Knowledge Integrated Plausible Motion Forecasting. IV 2024.

PDF Cite DOI

(2022). Plausibility Verification for 3d Object Detectors Using Energy-based Optimization. ECCVW 2022.

PDF Cite Video

(2022). Knowledge Augmented Machine Learning with Applications in Autonomous Driving.

PDF Cite

Contact